Affinity purification of Csk protein tyrosine kinase based on its catalytic requirement for divalent metal cations.

Protein tyrosine kinase Csk requires two Mg2+ ions for activity: one magnesium is part of the ATP-Mg complex, and the second free Mg2+ ion is required as an essential activator. Zn2+ can bind to this site to replace Mg2+, which inhibits Csk kinase activity. The binding is reversible and removal of Zn2+ results in an active Csk apoenzyme. In this communication, we report that this tight binding can be used as a mechanism for affinity purification of Csk. When bacterial cell lysate containing overexpressed GST-Csk was applied to a column of Zn2+-iminodiacetic acid immobilized to agarose, Csk was specifically retained by the column. Since the binding of Csk to Zn2+ is not affected by up to 200 mM NaCl, high ionic strength conditions were used in the purification procedure, minimizing nonspecific binding due to ionic interactions. Washing the column with 200 mM NaCl and 50 mM imidazole removed virtually all other proteins from the column while Csk remained bound. The retained Csk enzyme was eluted with 1 M imidazole. The 1 M imidazole-eluted fraction contained pure Csk that had a specific activity similar to the enzyme purified by a glutathione-agarose affinity column.

Cyclic peptides incorporating 4-carboxyphenylalanine and phosphotyrosine are potent inhibitors of pp60(c-)(src).

The protein tyrosine kinase, pp60(c-)(src), is involved in cellular signaling and is activated during mitosis and in various tumors. We have been employing cyclic decapeptides to identify the determinants for substrate binding and phosphorylation to develop inhibitors competitive with protein substrates of Src. A structure-activity study [McMurray, J. S., Budde, R. J. A., Ke, S., Obeyesekere, O. U., Wang, W., Ramdas, L., and Lewis, C. A. (1998) Arch. Biochem. Biophys. 355, 124] revealed that, at the position 3 residues C-terminal to the phosphorylated tyrosine (Y + 3), both glutamic acid and phenylalanine gave identical K(i), K(m), and V(max) values. We hypothesized that the area of Src that binds the Y + 3 residue contains either a positively charged lysine or an arginine, capable of ionic interactions with glutamic acid or cation-pi interactions with phenylalanine. To test this hypothesis, a series of phenylalanine analogues were substituted at position 7 (the Y + 3 residue) in cyclo(Asp(1)-Asn(2)-Glu(3)-Tyr(4)-Ala(5)-Phe(6)-Phe(7)-Gln(8)-D-Phe(9 )-Pro(10)). Of these, 4-carboxyphenylalanine (4-Cpa) and phosphotyrosine resulted in high affinity peptides exhibiting K(i) values of 0.85 and 1.1 microM, respectively, 180- and 130-fold increases in potency over the parent cyclic peptide (K(i) = 150 microM). These peptides were noncompetitive with respect to ATP and competitive against the phosphate-accepting substrate, polyGlu(4)Tyr. The truncated cyclic peptide, cyclo(Phe-4-Cpa-Gln-D-Phe-Pro-Asp-Aca) (Aca = epsilon-aminocaproic acid), which did not contain tyrosine, was also a competitive inhibitor with a K(i) value of 24 microM. We conclude that these cyclic peptides bind to a positively charged area that is near the phosphate transfer region of the active site of Src but does not necessarily include the tyrosine-binding pocket. Furthermore, the 4-Cpa-containing cyclic decapeptide shows remarkable selectivity in the inhibition of Src versus the src family members Yes and Lck, as well as other protein tyrosine kinases, Ser/Thr kinases, and other ATP-utilizing enzymes.

Benzodiazepine compounds as inhibitors of the src protein tyrosine kinase: screening of a combinatorial library of 1,4-benzodiazepines.

We screened 1680 spatially separated compounds of a diverse combinatorial library of 1,4-benzodiazepines for their ability to inhibit the kinase activity of protein tyrosine kinases Src, Yes, Abl, Lck, Csk, and fibroblast growth factor receptor. This screening yielded novel ligands for the protein tyrosine kinase Src. In the 1, 4-benzodiazepine-2-one scaffold, the preferred substituent at position R(1) was 4-hydroxyphenylmethyl or a 3-indolemethyl derived from a tyrosine or tyrptophan used in building the benzodiazepine, while the substituent at R(2), introduced by alkylating agents, was preferably aromatic in nature. The preferred ring structure introduced on the bicyclic ring of the scaffold by acid chlorides was a p-hydroxy phenyl group. The lead compound, designated as N-L-Yaa, has a L-4-hydroxyphenylmethyl ring at R(1) and a biphenylmethyl substituent at R(2). The compound has an IC(50) of 73 microM against Src, 2- to 6-fold lower than against other protein tyrosine kinases and >10-fold lower than against other nucleotide-utilizing enzymes. The mechanism of binding of N-L-Yaa to Src is mixed against the peptidic substrate with a K(i) of 35 microM and noncompetitive against ATP-Mg with a K(i) of 17 microM. Multiple inhibition analysis of the lead compound in the presence of other competitive inhibitors demonstrated that the binding of the lead compound is nonexclusive to the other competitive inhibitor. The inhibitor was found to be nontoxic to the AFB-13-human fibroblasts cells and inhibited the colony formation of HT-29 colon adenocarcinoma cells that are dependent on Src activity.

Mutations in the N-terminal regulatory region reduce the catalytic activity of Csk, but do not affect its recognition of Src.

In addition to the C-terminal catalytic domain, Csk is a protein tyrosine kinase that has an N-terminal regulatory region that contains SH3 and SH2 domains. The role this region plays relative to the function of the catalytic domain is not clear. To study its role, we introduced either deletion or site-specific mutations within this region and analyzed the effect of such mutations on the catalytic activity of Csk and its ability to phosphorylate/inactivate Src protein tyrosine kinase, its physiological substrate in the cell. Deletion of the SH3 domain and the SH2 domain resulted in reductions of kinase activity by 70 and 96%, respectively. Mutations within the SH2 domain that abolished its ability to bind phosphotyrosine did not result in a significant loss of kinase activity. Mutation of Ser78 to Asp, located between the SH3 and the SH2 domains, resulted in a reduction of over 90% of the catalytic activity. The reduction in specific activity is not the result of any apparent physical instability of the mutants. Kinetic analyses indicate that the mutations did not affect the Km values for ATP-Mg or the polypeptide substrate. The ability of the mutants to phosphorylate and inactivate Src is directly correlated to their kinase activity. These results indicate that the regulatory region is important in optimizing the kinase activity of the catalytic domain, but apparently plays no direct or specific role in substrate recognition.

N-myristoylation of a peptide substrate for Src converts it into an apparent slow-binding bisubstrate-type inhibitor.

The conversion of a peptide substrate to a potent inhibitor by chemical modification is a promising approach in the development of inhibitors for protein tyrosine kinases. N-acylation of the synthetic peptide substrate NH2-Glu-Phe-Leu-Tyr-Gly-Val-Phe-Asp-CONH2 (EFLYGVFD) resulted in synergistic inhibition of Src protein kinase activity that was greater than the inhibition by either free peptide and/or free acyl group. Synergistic inhibition was dependent upon the peptide sequence and the length of the acyl chain. The minimum length of the fatty acyl chain to synergistically inhibit Src was a lauryl (C11H23CO) group. N-myristoylated EFLYGVFD (myr-EFLYGVFD) inhibited the phosphorylation of poly E4Y by Src with an apparent Ki of 3 microm, whereas EFLYGVFD and myristic acid inhibited with Ki values of 260 and 35 microm, respectively. The nonacylated EFLYGVFD was a substrate for Src with Km and Vmax values of 100 microm and 400 nmol/min/mg protein, respectively. However, upon myristoylation, the peptide was no longer a substrate for Src. Both the acylated and non-acylated peptides were competitive inhibitors against the substrate poly E4Y. The non-acylated free peptide showed mixed inhibition against ATP while the myristoylated peptide was competitive against ATP. Myristic acid was uncompetitive against poly E4Y and competitive against ATP. Further analysis indicated that the myristoylated peptide acted as a reversible slow-binding inhibitor with two binding sites on Src. The myristoylated 8-mer peptide was reduced in size to a myristoylated 3-mer without losing the affinity or characteristics of a bisubstrate-type inhibitor. The conversion of a classical reversible inhibitor to a reversible slow-binding multisubstrate analogue has improved the potency of inhibition by the peptide.

Substitution studies of the second divalent metal cation requirement of protein tyrosine kinase CSK.

In addition to a magnesium ion needed to form the ATP-Mg complex, we have previously determined that at least one more free Mg2+ ion is essential for the activation of the protein tyrosine kinase, Csk [Sun, G., and Budde, R. J. A. (1997) Biochemistry 36, 2139-2146]. In this paper, we report that several divalent metal cations, such as Mn2+, Co2+, Ni2+, and Zn2+ bind to the second Mg2+-binding site of Csk with up to 13200-fold higher affinity than Mg2+. This finding enabled us to substitute the free Mg2+ at this site with Mn2+, Co2+, Ni2+, or Zn2+ while keeping ATP saturated with Mg2+ to study the role of the free metal cation in Csk catalysis. Substitution by these divalent metal cations resulted in varied levels of Csk activity, with Mn2+ even more effective than Mg2+. Co2+ and Ni2+ supports reduced levels of Csk activity compared to Mg2+. Zn2+ has the highest affinity for the second Mg2+-binding site of Csk at 0.65 microM, but supports no kinase activity, acting as a dead-end inhibitor. The inhibition by Zn2+ is reversible and competitive against free Mg2+, noncompetitive against ATP-Mg, and mixed against the phosphate accepting substrate, polyE4Y, significantly increasing the affinity for this substrate. Substitution of the free Mg2+ with Mn2+, Co2+, or Ni2+ also results in lower Km values for the peptide substrate. These results suggest that the divalent metal activator is an important element in determining the affinity between Csk and the phosphate-accepting substrate.

Autophosphorylation of Src and Yes blocks their inactivation by Csk phosphorylation.

Csk phosphorylates Src family protein tyrosine kinases on a tyrosine residue near their C-terminus and downregulates their activity. We previously observed that this regulation requires a stoichiometric ratio of Csk:Src in a time-independent manner. In this report we examined this unusual kinetic behavior and found it to be caused by Src autophosphorylation. First, pre-incubation of Src with ATP-Mg led to time-dependent autophosphorylation of Src, activation of its kinase activity and loss of its ability to be inactivated by Csk. However, the autophosphorylated Src can still be phosphorylated by Csk. The SH2 binding site for phospho-Tyr of this hyperactive and doubly phosphorylated form of Src is not accessible. Second, dephosphorylation of autophosphorylated Src by protein tyrosine phosphatase 1B allowed Src to be inactivated by Csk. Third, protein tyrosine phosphatase 1B preferentially dephosphorylates the Src autophosphorylation site and allows for Src regulation by Csk. Finally, Yes, another member of the Src family, was also only partially inactivated when a sub-stoichiometric amount of Csk was used. Mutation of the tyrosine autophosphorylation site of Yes to a phenylalanine resulted in a mutant Yes enzyme that can be fully inactivated by a sub-stoichiometric amount of Csk in a time-dependent manner. These results demonstrate that Csk phosphorylation inactivates Src and Yes only when they are not previously autophosphorylated and Src autophosphorylation can block the inactivation by Csk phosphorylation. This conclusion suggests a dynamic model for the regulation of the Src family protein tyrosine kinases, which is discussed in the context of previously reported observations on the regulation of Src family protein tyrosine kinases.

Cyclic peptides as probes of the substrate binding site of the cytosolic tyrosine kinase, pp60c-src.

A series of 48 cyclic peptides based on the amino acid sequence surrounding the autophosphorylation site of pp60(c-src) was synthesized and each was tested as both a substrate and an inhibitor of this protein tyrosine kinase. Starting with cyclo(Asp1-Asn2-Gln3-Tyr4-Ala5-Ala6-Arg7-Gln8-d- Phe9-Pro10) a six-amino-acid survey was performed at positions 1 through 8 to determine which positions were critical for affinity and phosphorylation and which amino acids produced the greatest activity. Our survey found that Arg7 was detrimental for binding and phosphorylation and that aromatic residues were preferred at this position. Further increases in affinity were obtained with hydrophobic residues at position 6 with the optimum for both affinity and phosphorylation being Phe. Changes on the "amino-terminal" side of Tyr4 resulted in reduced Vmax values, illustrating the requirement for acidic residues in peptidic tyrosine kinase substrates. The result of the survey was cyclo(Asp1-Asn2-Gln3-Tyr4-Ala5-Phe6-Phe7-Gln8-d-Phe 9-Pro10). The change of residues 6 and 7 resulted in a 42-fold increase in affinity and no increase in Vmax. As a substrate, this peptide displayed Michaelis-Menten kinetics at saturating ATP conditions. As an inhibitor, mixed inhibition was observed. A linear version of this peptide was 13-fold less potent an inhibitor than the cyclic peptide.

The instability of polyhydroxylated aromatic protein tyrosine kinase inhibitors in the presence of manganese.

Inhibition of the tyrosine kinase activity of Src by forty-three different compounds from five chemical families (cinnamic acid, salicylic acid, phenol, coumarin and flavonoid derivatives) representing plant and microbial secondary metabolites were studied in the presence of MgCl2 versus MnCl2. Within each chemical family, compounds containing multiple hydroxyl substituents demonstrated the greatest inhibitor potency. The ortho-substituted dihydroxy compounds were the most inhibitory. Except for the flavonoids, inhibition was higher in the presence of manganese compared to that observed with magnesium. UV-Vis spectra, HPLC, and mass spectrometric analyses demonstrate that manganese catalyzed the oxidation of these compounds. The general instability of such compounds, especially in the presence of manganese, and the associated problems it causes in the use of such compounds for developing selective protein tyrosine kinase inhibitors, is discussed.

Expression, purification, and initial characterization of human Yes protein tyrosine kinase from a bacterial expression system.

Protein tyrosine kinase Yes is a cellular homolog of v-Yes, the oncogenic protein product of avian sarcoma virus Y73. Yes is a member of the Src family and its activation has been associated with several types of human cancer. Human Yes has not been previously characterized enzymatically. To carry out biochemical characterizations of this enzyme, we expressed it as a fusion protein with glutathione S-transferase in Escherichia coli, to allow purification in a single step. The affinity-purified GST-Yes has a specific activity of 1.3 nmol min-1 mg-1 with polyE4Y as substrate and Km values of 100 microg ml-1 for polyE4Y and 70 microM for ATP-Mg. The enzyme has a preference for magnesium over manganese ion for maximal activity. The divalent metal cation serves two essential functions for the activity of Yes: one as a part of the phosphate-donating substrate ATP-Mg and the other as an essential activator. The enzyme undergoes autophosphorylation without apparent activation. Finally, we show that the enzyme is inactivated by incubation with protein tyrosine kinase Csk in an ATP-Mg-dependent manner, indicating that cellular Yes can be regulated by Csk phosphorylation. These represent the first biochemical characterization of human Yes protein tyrosine kinase.

Csk phosphorylation and inactivation in vitro by the cAMP-dependent protein kinase.

Csk is a protein tyrosine kinase that phosphorylates other protein tyrosine kinases of the Src family and down-regulates their activities. It is not known how Csk is regulated. We investigated the possibility that Csk is regulated through phosphorylation by examining if Csk can serve as an in vitro substrate for a panel of protein kinases. We found that Csk was phosphorylated by the cAMP-dependent protein kinase (PKA), but not by protein kinase C, Src, or the fibroblast growth factor receptor kinase. Csk phosphorylation in vitro by PKA is on a serine residue(s) and can reach a stoichiometry of approximately 0.6 mol phosphate per mole of enzyme. Furthermore, incubation with PKA in the presence of ATP and magnesium ion results in a time-dependent decrease in Csk kinase activity. A six-fold decrease in Csk activity (expressed as Vmax/Km ratio) was achieved due to a threefold increase in its Km and a twofold decrease in its Vmax value within 1 h of incubation with the catalytic subunit of PKA and ATP-Mg. Both phosphorylation and inactivation by PKA were blocked by a PKA-specific inhibitor. Csk mutants with a deleted SH2 or SH3 domain retained their ability to be phosphorylated and inactivated by PKA, indicating that the phosphorylation site is located within the catalytic domain. These studies suggest that the cAMP-dependent protein kinase can regulate Csk activity.

Requirement for an additional divalent metal cation to activate protein tyrosine kinases.

In addition to the magnesium ion needed to form the true phosphate-donating substrate (ATP-Mg complex), we have determined that at least one additional Mg2+ ion is essential for the activation of protein tyrosine kinases. This activation was investigated in detail using purified Csk, Src, and the fibroblast growth factor receptor kinase, which led to the following conclusions. (1) The catalytic activity of these kinases is dependent on the Mg2+ concentration present in the assay, approaching saturation at 5-8 mM MgCl2, while ATP was saturated at approximately 1 mM MgCl2. (2) Extrapolation to zero free Mg2+ at a constant ATP-Mg concentration predicts zero activity, suggesting that free magnesium ion in excess of that needed to bind to ATP is essential for the activation of these enzymes. (3) The free magnesium ion activates Csk and Src kinase activity by increasing the Vmax but does not change their apparent Km(ATP-Mg). In contrast, the free magnesium ion activates the fibroblast growth factor receptor kinase activity by increasing its Vmax and decreasing its apparent Km(ATP-Mg). These and previous studies with the insulin receptor tyrosine kinase suggest that receptor-type protein tyrosine kinases respond to the concentration of free Mg2+ differently than soluble protein tyrosine kinases. (4) With the phosphate-accepting substrate as the variable ligand, increases in the concentration of free Mg2+ resulted in increases in the apparent Vmax for all tyrosine kinases examined, but the apparent Km response is dependent on the enzyme and the substrate used. While these studies do not pinpoint a single kinetic mechanism, they do suggest that additional magnesium ion(s) is(are) an essential activator for protein tyrosine kinases in addition to being a part of the ATP-Mg complex. The difference among protein tyrosine kinases in their kinetic response to the additional divalent metal cation and the potential biological significance of such are discussed.

Tight-binding inhibitory sequences against pp60(c-src) identified using a random 15-amino-acid peptide library.

A bacteriophage peptide library containing a random 15-amino-acid insert was screened for identification of peptide sequence(s) that bind pp60(c-src). Sequencing the random insert from more than 100 virions indicated that more than 60% of the phage virions that bound to this enzyme contained a GXXG sequence motif in which X was frequently a hydrophobic residue. The GXXG sequence was often repeated as GXXGXXG. Two nonameric peptides were synthesized to determine whether or not the peptide inhibits pp60(c-src) tyrosine kinase activity and the importance of the glycine residues within this sequence. The peptide containing glycine had a Ki of 24 microM, whereas replacing the glycines with proline increased the Ki value to 3.1 mM.

The C-terminal Src kinase (Csk) is widely expressed, active in HT-29 cells that contain activated Src, and its expression is downregulated in butyrate-treated SW620 cells.

Csk expression was examined in human glioma, breast and colon cell lines, along with murine tissues. Relative to colon and glioma cell lines, Csk expression was highest in the breast cell lines. Relative to brain tissue, expression was 100-fold higher in the heart, kidney, liver, lung and spleen. Subcellular fractionation and cellular immunostaining indicated that it was localized in the cytosolic fraction. The expression of Csk was downregulated upon chemical-induced differentiation of SW620 human colon cells by treatment with sodium butyrate. Its expression localization and activity relative to Src are discussed.

A synthetic peptidic substrate of minimal size and semioptimal sequence for the protein tyrosine kinase pp60c-src.

We used a novel approach to determine the minimal size and semioptimal sequence of a peptide to serve as an inhibitor and/or substrate for the protein tyrosine kinase pp60c-src. The preferred amino acids surrounding tyrosine were determined by a systematic study in which we increased the length of a series of linear peptides starting from the tripeptide EYG. Using an iterative cycle, the size of the peptide was increased one residue at a time, first at the amino terminus and then at the carboxy terminus. A series of six analogs were synthesized at each position and assayed as inhibitors and substrates. The amino acids G, A, L, F, E, and K were used to semioptimize each position. The tripeptide EYG was not a substrate nor an efficient inhibitor. With increasing size of the peptide, the Ki decreased from 10.0 to 0.10 mM. The smallest peptide to serve as a substrate was a hexapeptide. The best overall peptide obtained from this method, EFEYAFF, had a Ki value of 0.13 mM with Km and Vmax values of 0.21 mM and 680 nmol/min/mg, respectively. Our best peptide was found to have higher substrate specificity than all other commerically available peptidic substrates for pp60c-src.

A tyrphostin-derived inhibitor of protein tyrosine kinases: isolation and characterization.

We recently reported that tyrphostin 23 (3,4-dihydroxybenzylidene malononitrile) is unstable in solution and that some of the degradation products are better inhibitors of the tyrosine kinase activity of Src and the EGF-receptor kinase than the parent compound itself (Ramdas et al., Cancer Res. 54, 867-868, 1994). In this study, the tyrphostin 23-derived compound designated P3, which is a more stable and potent protein tyrosine kinase inhibitor, was isolated. P3 was purified from oxidized tyrphostin 23 by solvent extraction, silica-gel flash chromatography, and reverse-phase high-pressure liquid chromatography. The physical characteristics of the isolated compound were determined and its chemical structure elucidated by 1H and 13C NMR spectroscopy. The proposed structure of this new inhibitor is that of a tyrphostin 23 dimer joined at the benzylidene carbon. P3 was evaluated in vitro as an inhibitor of four different protein tyrosine kinases (Src, Csk, EGF-receptor, and FGF-receptor) and two protein serine kinases (PK-A and PK-C). This compound exhibited the most inhibitory activity against Src with a Ki value of 6 microM and was less inhibitory toward the other protein kinases with Ki values ranging from 35 to 300 microM. P3 did not inhibit other nucleotide-utilizing enzymes such as lactate dehydrogenase and hexokinase. The growth and colony formation of HT-29 colon adenocarcinoma cells that contain activated Src was inhibited by P3 with an IC50 value of approximately 10 microM.

Production and characterization of monoclonal antibodies against the recombinant p50csk protein tyrosine kinase: a tool for signal transduction research.

p50csk is a protein tyrosine kinase (PTK) that has been reported to regulate the activity of other PTKs belonging to the src gene family. Several hybridoma clones that produce monoclonal antibodies (MAbs) directed against recombinant p50csk were established. Five of the clones were analyzed for their ability to recognize native and denatured p50csk protein after undergoing native and denaturing polyacrylamide gel electrophoresis followed by western blotting. In addition, the clones were tested for their ability to immunoprecipitate p50csk and yet maintain tyrosine kinase activity of antibody-bound p50csk. None of the clones cross-reacted with pp60c-src, a PTK that shares with p50csk the homologous SH1 catalytic domain and SH2 and SH3 regulatory domains. These MAbs can be used to study p50csk directly, and its role in regulating members of the src family.

Use of synthetic peptides and copolymers to study the substrate specificity and inhibition of the protein tyrosine kinase pp60c-src.

The ability of synthetic peptides and polypeptides to act as substrates and/or inhibitors of pp60c-src was examined. The random copolymer, poly(K4Y) had a threefold lower specificity than poly(E4Y). Peptides containing lysine vs. glutamate were also found to have a lower substrate specificity (Vmax:Km ratio). In order to assess the substrate specificity of acidic peptides, an assay protocol using DEAE-membranes was developed. Peptides containing a (YXE)5YXD motif (X = G, A, V, P, or norvaline) were tested as inhibitors and substrates of pp60c-src. The glycine-containing peptide was the best substrate having a specificity 16,000-fold higher than 5Val-angiotensin II, the most commonly used peptide substrate. Most of the peptides, except for the proline containing peptide, had Ki values of 20-100 microM. In a series of (XGE)5XGD peptides, where X = Y or F, tyrosine at position 10 was found to be the preferred site for accepting a phosphate. Analogs in which the glycine was replaced with alanine indicated that loss of flexibility around position 10 was detrimental to substrate specificity. Results suggest that conformational requirements of the peptides tested was important and substrate specificity was a more sensitive parameter than binding as measured by Ki values.